Structural Analyses of the Midshaft in an Isolated Femur from Koobi Fora, Kenya: Implications for Taxonomic Identity

Abstract

Reconstructing hominin evolution is dependent on our capacity to securely and accurately allocate fossil hominin material to an appropriate taxon. While taxonomic assignments are traditionally based on craniodental morphology, structural analyses of unassociated hominin long bones have provided a means to deduce the taxonomic identity of isolated postcranial remains based on morphological comparisons with corresponding elements from craniodentally associated material. This study examines cross-sectional geometric properties in the mid-diaphyseal section in KNM-ER 1592, an unassociated femur from the KBS Member of the Koobi Fora Formation, Kenya. Hominin taxonomic diversity throughout this member has hindered attempts to taxonomically place KNM-ER 1592 based on stratigraphic location alone. The aim of this study is to infer the taxonomic identity of this femur based on morphometric comparisons with femora assigned to Australopithecus, Paranthropus, and Homo. Relationships between cortical area and total area, and differential bending rigidity along the anatomical planes of the midshaft femur were examined. Torsional and average bending strength in KNM-ER 1592 was compared to distributions in australopith, non-habilis early Homo, and H. naledi groups to further assess its midshaft structure in comparative context. The overall mid-diaphyseal femoral morphology in KNM-ER 1592 fits comfortably with patterns observed in non-habilis early Homo Pleistocene femora. KNM-ER 1592 shows an increase in cortical area for a given total area (i.e., relatively thick cortical bone) in the midshaft region like most pre-modern hominins. However, it displays a greater mediolateral relative to anteroposterior expanded mid-diaphyseal morphology characteristic of H. erectus and “erectus-like” femora, and its torsional and average bending strength further demonstrates its affinity with non-habilis early Homo. Based on the structural analyses of the mid-diaphyseal region, a tentative assignment of KNM-ER 1592 to genus Homo sp. indet. is supported. The alignment of KNM-ER 1592 with Homo is further supported when its overall cross-sectional size is also considered. Overlap among australopith and Homo samples at the smaller and intermediate size ranges is observed in both bivariate comparisons, but australopith and Homo samples do not overlap at the larger end of the size range where KNM-ER 1592 consistently falls with KNM-ER 803a and 1807. The former has been craniodentally aligned with H. erectus and the latter has been morphometrically aligned with H. erectus and “erectus-like” femora. Biomechanical analyses provide an avenue for which to test hypotheses about the taxonomic identity of isolated postcranial material based on morphological comparisons with material craniodentally attributed to a specific taxon.